Cognitive Multi-User MIMO downlink with mixed feedback/location based Gaussian CSIT

The Multi-User MIMO downlink or MIMO Broadcast Channel (BC) formulation is relevant for cell center users. Whereas multiple receive antennas do not allow to increase the total number of streams (or degrees of freedom (DoF)) in the BC, they allow the sharing of zero-forcing (ZF) between transmitter and receivers so that a secondary base station (SBS) can serve its secondary users (SU) while ZF beamforming (BF) to primary users (PU). Channel State Information at the Transmitter (CSIT), which is crucial in multi-user systems, is always imperfect in practice, especially for the SBS-PU link. We consider mean and covariance Gaussian partial CSIT, and the special case of a (possibly location based) MIMO Ricean channel model. In this paper we focus on the optimization of beamformers for the secondary expected weighted sum rate (EWSR) under expected PU interference power constraints. We apply a perfect CSI technique, based on a difference of convex functions approach, to a number of deterministic approximations of the EWSR, involving the Massive MIMO limit (large number of transmit antennas), Massive MIMO with a second-order refinement, and the large MIMO limit (both large transmit and receive antenna numbers).